ada/5907: The Ada front end lacks a proper manual

[Florian Weimer]

The following reply was made to PR ada/5907; it has been noted by GNATS.

From: Florian Weimer <fw@deneb.enyo.de>
To: "Joseph S. Myers" <jsm28@cam.ac.uk>
Cc: <gcc-gnats@gcc.gnu.org>,  <gcc-bugs@gcc.gnu.org>,  <brosgol@gnat.com>
Subject: Re: ada/5907: The Ada front end lacks a proper manual
Date: Sun, 10 Mar 2002 15:14:48 +0100

 "Joseph S. Myers" <jsm28@cam.ac.uk> writes:
 
 > On Sun, 10 Mar 2002, Florian Weimer wrote:
 >
 >> The VMS version requires substantial postprocessing (words are
 >> replaced globally, file names are rewritten).  It is not possible to
 >> express this in Texinfo.
 >
 > Why not?  If the standard manual needs the word "foo", where the VMS 
 > manual needs "bar", why shouldn't a macro @foo{} (with different 
 > definitions in the two cases) work?
 
 We would need a couple of hundered macros, I think.  For example, the
 preprocessing rewrites all strings like "hello.adb" to "HELLO.ADB",
 "hello.o" to "HELLO.OBJ", and so on.  If you had to write
 "@helloadb{}" and "@helloo{}", reading the sources would become rather
 difficult.
 
 > I don't think however the VMS way of building the manual would be of
 > relevance to the FSF sources - I think having just one version, covering
 > all systems, is generally preferred for GNU manuals.  (If it were done 
 > simply by a VMS Texinfo conditional, it might still make sense not to 
 > define that conditional when building the GNU manual on VMS.)
 
 Of course, the preprocessed manual could be committed to the CVS. But
 I doubt that this is a reasonable approach.
 
 >> bit clumsy, and the TeX implementation of Texinfo doesn't handle
 >> extensive use of conditional processing and macros very well.  On the
 >
 > What are the problems?
 
 Macro expansion in index entries causes problems, and so do some kind
 of conditionals (but these problems are more related to makeinfo,
 IIRC).